Paradoxically, the morphogen retinoic acid, which is required for early tissue interactions in the embryo, is deleterious to in utero development at pharmacological levels. Previous studies have demonstrated that the cynomolgus macaque is highly sensitive to the adverse effects of this retinoid on the developing fetus. Further, the similarities with humans in the pharmacokinetics, malformation syndrome, threshold dose and early sensitive period for induction of defects following exposure to 13-cis retinoic acid (accutane) make this macaque an ideal species for risk assessment of additional retinoids. Several experiments were done to further define the most effective dosing regimen for studies involving the parent compound, vitamin a (retinol). The results showed that the double dose on gd26-27 (after single exposure on gd12-25) was not essential to teratogenicity, and that multiple daily doses on gd26-27 failed to induce any manifestation of abnormal development. Similarly, exposure during forelimb development (gd20-30) was unsuccessful in inducing defects of these tissues, although defects in several retinoid target organs (i.E. Cerebellum, internal ear) were present, verifying the teratogenic potency of the dose. The results confirm that the lowest observed effect level (loel) in macaques is 2.5 Rather than 5.0 Times greater than that observed in human pregnancies since the retinoid syndrome was demonstrated following single daily doses of 2.5 Mg/kg. Pharmacokinetic parameters for the administered drug 13-cis retinoic acid and its main metabolites reached maximal levels (CMAX) by six and two hours on gd12 and 27, respectively, indicating rapid absorption. The AUC values showed slight increases from gd12 to gd27; no clear-cut trend in kinetic parameters was correlated with fetal outcome. The results support the continued use of the macaque, using a specific treatment schedule, as an appropriate model for screening retinoids developed for therapeutic purposes.